1p(1mpal)

여기 게시글에 바이럴 탔던 프롬프트&꿀팁 전부 모아둠 🚩어떤 프롬프트나 꿀팁을 엄선해 두었냐고? 1. 누가 봐도 와우 모먼트가 있을 것 2. 누가 써도 따라하기 쉬울 것 마음껏 베껴도 되고, 마음껏 응용해봐도 돼! 마케팅 업무에도, X 계정 수익화에도 도움 될 거 장담함👇

[프롬프트(웬만하면 응용을 추천!] Authentic 2000s–2010s Korean terrestrial sports-variety broadcast — a summer water-park special of a major Korean obstacle-survival show. TV sports coverage mixing ENG, crane, wire-cam and telephoto cameras; natural handheld shake, zooms, pans, tracking. Slightly soft early-HD image, broadcast noise, TV compression, realistic color. A special obstacle course in an outdoor water park: midsummer midday sun, blue sky, big water slides and palm trees, spectator stands, commentary booth, a roaring crowd. Audio: an excited Korean male sports caster commentates live throughout, in the exaggerated tone of 2000s Korean weekend variety. Korean crowd cheers and applause, buzzer, water, venue ambience. All voices in natural Korean. The main subject is locked entirely to @ image1. Her facial features, hairstyle, skin tone, body type, apparent age, expression, outfit and identity never change from start to finish; she never morphs into someone else; facial contours and body proportions hold. The outfit follows the reference image and stays the same, even soaked. 0–3s [Knee-up frontal shot] ENG challenger introduction. The @ image1 person stands at the start point, smiles brightly at the camera, waves, and shouts in Korean: "출발하겠습니다! 드림팀 화이팅!" Staying in place, she slams her palm on the red start buzzer on the post beside her; as it sounds, white smoke bursts from both sides with a sharp hiss, and she immediately grabs the overhead zipline rope handle with both hands. Applause and cheers erupt. 3–7s [High rear tracking shot → water-level side tracking shot] As she kicks off, cut to a tracking shot in the air right behind her: hanging from the handle, with the vast pool course, stands on both sides and water slides spread out below. Then cut to a side tracking shot low over the water, following her gliding across the surface against blue sky and cumulus clouds. At the end she lets go and lands on the yellow round landing platform. Rapid-fire commentary. 7–9s [Telephoto tracking shot] She charges into the row of large floating buoys, crossing them one step per buoy. She wobbles hard but spreads her arms and pushes through rhythmically. The stands roar; the caster's voice rises. 9–12s [Handheld zoom-in → water-surface fixed camera] She stops on the last buoy, looking up at the monkey bars over the water — an overhead horizontal ladder of metal rungs. A handheld camera slowly zooms in on her face at a slightly tilted angle; the commentary holds its breath. She exhales "후..." to steady herself, then leaps for the first rung — her fingertips only graze it and she plunges straight into the pool. On impact, cut to a camera fixed just above the water surface, engulfed by the splash: the frame fills with out-of-focus swirling blue water and white bubbles. Gasps and laughing cheers erupt. 12–15s [Water-surface low-angle fixed camera] Same camera. As the spray settles, her head rises slowly out of the water, wet hair draped like a curtain over her whole face. Laughter bursts out. She sweeps the hair back with both hands, revealing a beaming smile, droplets sparkling in the sun. Behind her: water slides, palm trees, blue sky. Smiling, she waves at the camera; the crowd applauds and the caster cheerfully wraps up. Video quality & style Faithful 2000s–2010s Korean sports-variety texture: real ENG/crane/wire-cam/telephoto coverage, natural zooms, pans, tracking, soft early-HD grain, broadcast noise, TV compression, live-action color. Real-time broadcast pacing throughout; no slow motion at any point. Top priority Maintain the @ image1 person's identity 100% — face, hairstyle, body type, skin tone, apparent age and outfit held to the very end. No morphing, facial distortion or body-shape change. No captions, logos, watermarks, subtitles or timer graphics; no real broadcaster or show trademarks. All commentary and crowd voices in natural Korean. Realistic broadcast quality, like a real Korean obstacle show on air, with natural sports-broadcast camera work.

천만 뷰 찍어버린 그라비아 모델 특집 출발 드림팀 역시 언어를 뛰어넘는 만국 공통의 취향이란.. @DomoAI_로 만듬! 준비물 1. 원하는 캐릭터 수영복 입히기 (가장 어려움) 2. 아래 프롬프트를 응용해서 각자의 코스 만들기 수영복 착장은 Seedream5.0 추천! 한국어 프롬프트는 타래에 적어놨어!

@Chengzilhy @YangOnchain 不，没关系。prompt本身并没有原创者。任何人都可以加以运用，我本人也经常分享自己对他人作品的改编成果。

2个10秒视频拼接出来的街角创意次元之旅 提示词在评论区 omni制作~

@quietcomp_ @get_artcraft 프레임이 조금 녹았지만 괜찮았다면 감샤샤샤 합니다

@impalementd @get_artcraft ㅋㅋㅋㅋㅋㅋㅋ 좋은데요 ?

리센느 프리티걸 챌린지 커버하는 방법 등장인물이 많아? 감정선이 필요해? 준비물 1. 원하는 틱톡 챌린지 다운 2. 커버 영상의 Depth 영상을 추출한다(클로드 해줘!) 3. 레퍼런스 비디오에 넣고 시댄스로 제작 제작할 때 사용한 프롬프트도 아래에 남길게! > Made with @get_artcraft

[프롬프트] Create a 15-second horizontal 16:9 cinematic magical-girl transformation video. The character's identity follows the uploaded reference image: keep her exact face, skin tone, eye color and eye shape, hair color, signature hair ornaments, jewelry, and color palette from the reference consistent throughout. The transformation replaces her original reference outfit with a completely new magical-girl costume. Render in soft cel shading with glossy eyes, luminous jewels, elegant sparkles, and flowing hair. 0:00–0:01.5 Extreme close-up of one eye and an earring. 85mm macro push-in. The earring flashes, jewel light reflects in her pupil, and her expression shifts from surprise to determination. 0:01.5–0:03.2 Medium shot. She brings both hands to her chest as light gathers between her palms. Her outfit, hair, and ornaments lift in a magical breeze. A circular aura in her signature colors forms around her. Slow dolly-in. 0:03.2–0:05.4 Low-angle full-body shot. She floats upward as her original outfit dissolves into glowing ribbons, jewel fragments, and shimmering particles. Use a classic 1980s–1990s magical-girl transformation effect: her body becomes a radiant sparkling silhouette in soft pastel light from her reference palette, with star glints, soft bloom, glossy highlights, and a clean luminous outline. Keep it elegant, stylized, and simplified. 0:05.4–0:07.2 Smooth clockwise orbit. Ribbons in her signature colors wrap around the glowing silhouette as a completely new costume assembles piece by piece: a frilled layered mini-dress in her palette, long gloves with jewel cuffs, knee-high magical boots, and a large chest bow. A heart-shaped gemstone locks into the center of the bow. Add sparkling flashes as each costume part locks into place. 0:07.2–0:08.8 Close-up. Her hair sweeps upward and re-forms into long, flowing twin tails with small round buns at the crown — classic 90s shoujo heroine styling — keeping her hair color from the reference. Jeweled hair ornaments flash as the twin tails settle. She opens one eye with a playful wink. Glitter trails, halo rings, and starbursts follow the hair motion. 0:08.8–0:11.0 Wide 24mm hero shot. She spins gracefully in midair in the completed magical-girl costume: frilled mini-dress, chest bow with heart jewel, long gloves, knee-high boots, and flowing back ribbons in her reference palette, clearly different from her original outfit. A luminous jewel sigil appears beneath her. Pastel light bursts outward. Subtle speed ramp during the spin. 0:11.0–0:13.0 She lands lightly on one foot inside the fading magical circle. Low-angle push-in. One hand extends forward while heart-shaped jewel energy condenses above her other palm into a short magical wand — a compact rod topped with a heart-shaped jewel in her signature colors. Hair and ribbons settle naturally. A soft impact ring and glitter splash spread on the floor. 0:13.0–0:15.0 Final medium close-up hero pose. She strikes a classic magical-girl finishing pose with the short wand: one hand sweeps it up beside her face, its jewel tip flaring with a starburst, while the other hand rests on her hip, smiling brightly toward the camera. A radiant jewel crest halo glows behind her with floating gems, star glints, and ribbon trails. Slow 50mm push-in and hold the final silhouette clearly. Style: retro 1990s shoujo magical-girl anime, hand-drawn cel animation, pastel jewel fantasy, crisp elegant outlines, soft airbrushed highlights, iridescent gemstones, sparkling star flares, graceful ribbon motion, classic 80s–90s glowing transformation silhouette, cinematic horizontal composition, readable silhouette. Avoid: face inconsistency, identity drift from the reference image, hair color changes, missing hair ornaments or jewels, broken anatomy, the original outfit reappearing, random props, extra characters, text, subtitles, logos, UI, photorealism, 3D rendering, shaky camera, heavy motion blur, dark horror lighting.

자캐로 레트로풍 마법소녀 변신! 시키는 방법 여러분의 자캐를 90년대 마법소녀로 바꿔보자! 실사체도 잘 나오는 거 확인함! 준비물 1. 자캐 이미지 편집 (화풍이 레트로일수록 좋음!) 2. 시댄스로 프롬프트랑 같이 넣고 제작하면 끝! 프롬프트는 아래에 남겼어! Made by @DomoAI_ #MadewithDomoAI

@clairejiyeonj @get_artcraft 하와와 미소녀 치어리더예요

@impalementd @get_artcraft 순간 직접 하신줄알고 당황;;

@Gorilla30735604 Yes, it's certainly not the easiest way, but it gives you more flexibility.

@impalementd kleinの方が簡単な気がするよ

원하는 댄스 커버 무한으로 찍어내는 방법 커버 영상 만들 때 저작권 이슈 터질 때 있지? 이걸로 간단 해결! 준비물 1. 마음에 든 15초 이내 동영상 저장 2. 영상의 Depth를 추출하는 사이트를 클로드로 제작 3. 만들어진 사이트에 영상을 넣고 Depth 영상을 추출 4. 하단 프롬프트처럼 넣고 제작!

@kaigooddev @get_artcraft 만들고 나서 실수한게 depth 뽑을 때 fps 낮게 세팅되어 있는걸 지금 알아서.. 광학 흐름 줬는데도 좀 삐걱거리네요 ㅠ

@impalementd @get_artcraft 깊이 정보 뽑아두면 포즈가 덜 무너지죠 ㅋㅋ 이게 실무에서 제일 먹히더라구요

[크로마키샷에 사용했던 프롬프트(감정선 보강)] 15-second continuous single-shot dance clip, vertical 9:16. No cuts. No scene transitions. Motion: @ Video1 is a depth-map dance reference containing TWO dancers. Reproduce both silhouettes, body poses and every motion beat one-to-one on the same timeline, with the same locked camera framing as @ Video1: the camera stays completely fixed for all 15 seconds — every framing change comes from the dancers themselves stepping toward and away from the lens, including the beat where the taller RIGHT dancer advances into a close chest-up distance and the finale where she dances large in the right foreground while the LEFT dancer dances behind her on the left. Do NOT move, zoom or reframe the camera. The screen-LEFT dancer is the woman from @ Image1 — long black hair in a high ponytail with a blue scrunchie, straight bangs, grey-brown eyes, white-and-blue cheerleader uniform: white collared halter crop top with a blue check panel and blue straps, bare midriff, white satin elbow-length gloves with a blue ribbon on one arm, white pleated miniskirt with blue pleats over black shorts, a fluffy yellow-and-white pom-pom gripped in each hand — held identical in every frame. The screen-RIGHT dancer is the woman from @ Image2 — long blonde hair in a high side ponytail with a black ribbon, a single ahoge strand, red eyes, black spiked choker, red buttoned crop top with a green bow and a white rabbit badge over a black diamond-pattern corset with gold eyelets, white detached puffy sleeves with black patch armbands, black layered ruffle miniskirt with a gold star belt badge, thigh straps and black thigh-high stockings with cutout lacing, a thick white tail accessory, chunky black platform boots — held identical in every frame. Style: each dancer keeps the exact rendering style of her own reference image — the LEFT dancer stays photorealistic live-action with natural skin and fabric texture exactly as @ Image1; the RIGHT dancer stays a high-quality 3DCG cel-shaded toon render with crisp clean outlines exactly as @ Image2 — the two styles coexisting in one frame. World: there is no location at all. Every pixel of the frame around the two dancers is one flat solid green color fill (#00FF00) — a pure 2D graphic color layer, the exact same green tone in every pixel edge to edge, top of frame identical to bottom of frame, as if the dancers were already keyed out and composited over a solid green card. The dancers receive no scene lighting: each keeps her own look straight from her reference image — same colors, same shading — with crisp clean silhouette edges against the flat green. Both dancers are already dancing from frame one — the LEFT dancer's ponytail, pleated skirt and pom-poms shaking on every beat, the RIGHT dancer's side ponytail, ruffle skirt and tail swinging with each move. Acting: the two dance with playful chemistry, like close friends goofing off — every few beats they trade quick sideways glances at each other, grinning, smirking, a raised eyebrow, a cheeky wink, mouthing along to the beat, breaking into a short laugh between moves — lively mischievous facial expressions the whole time, while their bodies keep following @ Video1's choreography exactly on every beat. The clip ends mid-move, flowing into the next motion. Avoid: camera movement, camera drift, zoom, pan, pull-back, a physical green-screen studio set, fabric screen wrinkles, wall-floor seam, horizon line, corner lines, gradient or vignette in the green, floor shadows, reflections, studio lighting, rim light, glow, green color cast on the characters, blank frozen expressions, face morphing, distorted hands, the two dancers merging or swapping positions, extra people, text, watermark, logo.

@Bugi952 유튜브는 좀 콘텐츠로서의 허들이 높은 느낌이예요 스토리가 명확해야 한달지

인스타는 빵빵 유튜브는 엉엉 무슨 차이일까요?

@Chengzilhy I guess I'll have to start churning out dance covers like a factory now.

@impalementd This is very practical.

@nowlovepan 그런 면에서 제미나이가 마케팅을 잘 한 느낌? 요즘 애들 관용어구가 "제미나이에 불어보니까~"인 것 같아서.,

이런 거 볼 때마다 느끼는 점이지만, AI 인플루언서들은 앞으로 더 크게 성장할 수 있는 최적의 환경이 열리는 것 같음. 정부 차원에서 '국산 AI 모델' 기반 대국민 서비스를 무료로 풀기 시작하면 일반 대중의 AI 친숙도가 지금이랑은 비교도 안 되게 높아질 수밖에 없음 기술 진입장벽은 낮아지고 대중성은 올라가니, 이 타이밍에 캐릭터성 잘 잡고 서사 쌓아 올리는 AI 인플루언서들이 시장을 큰 파이로 먹을 확률이 높다고 봄 진짜 기회는 준비된 사람한테 옵니다 지금 당장 결과가 안난다고 포기하지말고 꾸준히 가봅시다.

[사용했던 프롬프트] 15-second continuous single-shot summer beach dance video. No cuts. No scene transitions. Motion: @ video_1 is a depth-map dance reference — follow its choreography, timing and framing 1:1, matching every pose and beat exactly. Camera framing follows the reference video. Character: @ image_1 — exactly as the reference. Setting: bright summer beach at midday — golden sand, rolling turquoise surf, clear blue sky. Style: photorealistic, true camera texture, crisp high-clarity plate. LIGHT: natural bright sunlight, high-key, hard warm key from the sun, sparkling water highlights, soft sand bounce fill. Sun-drenched, never grey. She is already dancing from frame one; hair, skirt and sea breeze react naturally to the motion. Avoid: extra people, face morphing, distorted hands, low grey light, on-screen text, subtitles, watermark.

@0xMementoMori people just want to next bigcoin.. and yeah..

What used to "work" for Ponzis is changing. My new rule of thumb is to avoid any where the utility token equals the reward token (like UN3D). Or, if you get in early (roughly sub-$1M), you buy the token and dip early, but you don't buy a setup.

@ingi_erlingsson @ComfyUI @krea_ai @Alibaba_Wan @thesystms @suno Wow, very sesitive artwork..

nature vs nurture 🌺 made in @ComfyUI with @krea_ai 2 + @Alibaba_Wan 2.2 I2V/T2V/Vace + @thesystms HUD + @suno 🎹

@EHuanglu Wow, did you make all of this yourself again? It turns out the process is more complicated than I thought.

this dragon ball live theatre in japan is insane

좋은 건 나눠야해 Web3, 그러니까 코인으로 알게 됐지만 AI 영상 가장 열심히 하시는 분이 바로 임팔 님 평소에도 프롬프트 자주 올리시니 AI 영상 만드시는 분들은 팔로잉해두면 더 좋겠지요? 저도 임팔님 덕에 시댄스 구독 뽐뿌가 매번 오고 있는데 여유만 생기면 그냥 막 확...